2 backup jobs to be run concurrently

[hanneng]

I have 2 LTO drives in the 3582 Tape Library.



When I scheduled 2 backup jobs to be running at the same timing,

however, only 1 job run at a time, while the other backup job is on the q.



In order to save time, I needs both backup jobs run concurrently,

using both LTO drives .



How to configure that to happen.



Thank You

 

[may]

have a look on the



"RessourceUtilization" on the client and the "Max mount Point" option on the server!





by the way, do you have some free disk space ?? because you can define a Storage Pool Disk, and redirect all your backup on it when migrate them to your Drives !

 

[hanneng]

I have configured the following:



Client

RESOURCEUTILIZATION: 5



Server

Maximum Sessions: 25

Maximum Mount Points Allowed: 10



However 2 backup jobs (with same scheduled time) is still not running concurrently.



Please help



Thank You

 

[may]

on your client set the



RECOURCEUTILIZATION to 2 ( one for data flu and the other one for Meta data)



and on the server , go to your nodes configuration page and set as



Max Mount Point to only 1 (that will force the client to use only One drive a time)





hey just to be clear, is these two backup job you want to launch are on the same node ??

 

[itdrew]

may,



Is it advisable to keep the RECOURCEUTILIZATION in even numbers?



I ask because if you set RESOURCEUTILIZATION to say 4, is it true that you will get two data threads and two meta data threads? If you set it to six, then you get three and three?



Or do multiple data threads share the same meta data thread?

 

[may]

hello Itdrew,



u can set the RESOURCEUTILIZATION to what you want, if the Max mount point allowed on the server side for the client is just one (i am for tape backup) there will be just one drive granted to the node



the RESOURCEUTILIZATION is more importante for restore than backup mode



here is something wich will hel you better understand the RESOURCEUTILIZATION :



Default resourceutilization level allows up to two sessions with the server, one for querying the server and one for sending file data. Resourceutilization=5 permits up to four sessions (two for queries and two for sending data), while resourceutilization=10 permits up to eight sessions (four for queries and four for sending data) with the server. The relationship between Resourseutilization and the maximum number of sessions created is part of an internalized algorithm and, as such, is subject to change. This table lists the relationships between Resourceutilization values and the maximum sessions created. Producer sessions scan the client system for eligible files. The remaining sessions are consumer sessions and are used for data transfer. The threshold value affects how quickly new sessions are created.





hope it helped

 

[hanneng]

Thanks for your reply, May.



To answer your question, " is these two backup job you want to launch are on the same node ?? "



Yes. Both jobs are on the same node.

But is configured with different Policy Domain Name.



I will try your suggestion, and let check the outcome tomorrow.



thanks

 

[may]

hey,

that twist all the story,



be careful, is this concern the same mashine or the same node?



if you wanted to launch 2 backups same time on the same node, my solution won't work, you will have to increase the "Maw mount point" and the RESOURCEUTILIZATION, 'cause you' r going to have have several sessions.....

 

[hanneng]

Hi may,

thanks for your clarification.



I've confirmed that changing RECOURCEUTILIZATION to 2 and Max Mount Point to 1

doesn't resolve my problem.



I have 2 LTO drives as shown below:



tsm: TSM>q drive



Library Name Drive Name Device Type On-Line

------------ ------------ ----------- -------------------

3582LIB DRIVE0 LTO Yes

3582LIB DRIVE1 LTO Yes

------------------------------------------------------------------------------



However, APP_DAILY scheduled job will start after DATA_DAILY job completed.



tsm: TSM>q event * * begind=today-1 f=d



Policy Domain Name: SG_POLICYDOMAIN

Schedule Name: DATA_DAILY

Node Name: SG

Scheduled Start: 07/29/05 23:30:00

Actual Start: 07/29/05 23:30:03

Completed: 07/29/05 23:57:35

Status: Completed

Result: 0



Policy Domain Name: SG_POLICYDOMAIN

Schedule Name: APP_DAILY

Node Name: SG

Scheduled Start: 07/29/05 23:30:00

Actual Start: 07/29/05 23:57:35

Completed: 07/30/05 00:06:25

Status: Completed

Result: 0



--------------------------------------------------



So, my question is How to make both scheduled jobs run concurrently to save time ??



Thank YOu

 

[may]

hey,



how did you differe the 2 backup in their allocation ??



do you have two scheduler ?? If yes why not have two "Dsm.opt" ??



it will resolve all your problem

 

[dmdbt]

attention!



the option "-optfile=<myopt> is NOT supported as schedule option but command line !!!



dmdbt

 

[may]

you are right,



but you can use two different scheduler with two different dsm.opt....

 

[dmdbt]

yes of course, this is similar to command line



But be sure that you don't define collocation! If collocation defined, job 2 ist waiting for the tape used by job1



dmdbt

 

[saleemy]

hi



when you mention 2 backup jobs, does it mean each job backup different drives/shares/folders/files or both jobs should be able to backup the same



if you are backing up different data for each job then best way is to creat two nodes with different names, two dsm.opt files(put domain=respective drives, resource utilization=1), two schedulers each pointing to its own opt file

on server side you need to say max mount point=1 for each node and collocation=yes



hope this helps

mohamed

 

[tsmadm]

The best way to do this is to use a different node name for each backup schedule. TSM will only run one schedule at a time for each node. This is because a node can not access the server when it already has a session established.



This configuration is like a TDP backup.

 

[tsmadm]

The best way to do this is to use a different node name for each backup schedule. TSM will only run one schedule at a time for each node. This is because a node can not access the server when it already has a session established.



This configuration is like a TDP backup.

 

[itdrew]

It will work using a different node name. You just need to document which data is backed up under which node name.